IAB’s updated AI Transparency and Disclosure Framework says U.S. advertisers can use a standardized sparkle icon or a clear text label when AI materially changes what consumers see, hear or interact with.
The ad industry wanted AI to produce more creative, faster and cheaper. It’s getting that. It also inherited a new question for every asset moving between a brand, agency, ad server, platform and publisher:
Who’s responsible for telling the audience what’s real?
The Label Depends on What AI Did
AI can clean up audio, remove an object, write alternate copy, generate a background, clone a voice, create a spokesperson or place a real person in a situation that never happened.
IAB’s framework focuses disclosure on generated material that could confuse consumers about authenticity, identity or representation. That includes realistic prompt-generated images and video, certain synthetic voices, digital twins, synthetic avatars and AI assistants that could be mistaken for people.
Routine post-production, standard audio enhancement, background music, generic synthetic voices and clearly stylized characters generally won’t trigger a label automatically.
Labeling every AI-assisted edit would make the disclosure meaningless within a week. Consumers would see the icon on nearly everything and stop seeing it at all.
The Sparkle Icon Creates an Ownership Problem
Every label needs an owner.
A brand approves a campaign. An agency produces dozens of versions. A generative tool creates alternate backgrounds and voices. A dynamic creative platform assembles audience-specific variations. An ad server distributes them across formats. Publishers and streaming platforms render the final assets.
Someone has to know what the AI changed.
Someone has to determine whether that change requires disclosure.
Someone has to preserve the disclosure as the creative gets resized, localized, compressed and delivered.
A generic policy sitting with the legal department won’t handle that. The decision has to travel with the asset.
Brands will need records showing which tools were used, whether a real person’s likeness or voice appears, what was generated and who approved the final version. Agencies will need a repeatable review process. Platforms will need to render the label correctly.
The sparkle icon is the part consumers see. The expensive part is everything behind it.
CTV Is Built to Lose This Information
CTV creative can pass through a brand, agency, demand-side platform, ad server, publisher and device manufacturer before reaching a viewer.
Each handoff creates another chance for context to disappear.
A label placed safely inside one version can get cropped out of another. Text can become unreadable on a different screen. An agency can approve the original asset while a platform generates additional variations after trafficking. Metadata can get dropped as files move between systems.
The same fragmentation that makes transparency difficult across CTV advertising will make AI disclosure harder to enforce consistently.
The ad needs more than a visible icon. It needs an attached record describing how AI was used and which disclosure decision was made.
Regulators Have Already Started the Clock
IAB’s first disclosure framework arrived in January. Version 2 comes as governments begin turning AI transparency into enforceable rules.
New York’s synthetic-performer law took effect in June. California’s AI metadata and labeling requirements became effective in August. South Korea has introduced AI-labeling mandates. Article 50 of the European Union’s AI Act now requires disclosure for certain AI-generated content and deepfakes.
Those rules don’t all use the same thresholds, language or label format. The EU hasn’t finalized a common icon. U.S. requirements can vary by state and type of synthetic content.
A global campaign may therefore need different treatments depending on where it runs.
IAB is giving the industry a shared starting point. Brands and agencies still have to map that framework against the laws governing each market.
Disclosure adds classification, documentation, review and jurisdiction-specific requirements to every creative variation.
Disclosure Could Become Proof of Competence
IAB research found that more than half of surveyed Gen Z and Millennial consumers wanted brands to disclose fully AI-generated advertising or the use of AI imagery and video. Seventy-three percent said clear disclosure would increase or have no effect on their likelihood of buying the product.
A brand that hides synthetic content or can’t explain how it was produced creates a trust problem.
A consistent label can signal that the advertiser knows what its production systems are doing. It shows that the brand tracked the asset, reviewed the creative and made a deliberate disclosure decision.
The Streaming Wars Take
AI makes advertising production faster and creates a chain of responsibility across brands, agencies, platforms and publishers.
IAB’s framework gives the industry a practical rule: disclose AI when it materially changes authenticity, identity or representation in a way that could mislead consumers.
AI ads are getting a nutrition label. The companies serving them need to know what’s inside.
The Streaming Wars is intentionally ad-free
We don’t run display ads. Not because we can’t, but because we don’t believe in them.
They interrupt the reading experience. They cheapen the work. And they burn advertisers’ money on impressions nobody actually wants.
So we chose a different model.
We say the things people in this industry are already thinking but don’t say out loud. We connect the dots beyond the headline and focus on explaining why things matter to the people working in this business.
If you believe industry coverage can exist without clutter and interruption, you can support it here → SUPPORT TSW.
Support is optional. But it directly funds research and continued coverage — and helps prove this model can work.
Support TSW →
